luminarforge.
Pricing Compare Trust Log in Start free
The pitch desk — AI suggestions

The blank page, abolished.

Every morning the desk deals pitch slips from the client’s own brand brain — topic, platform, format, and the reason it’s on the pile. Take one and it lands on the calendar briefed. Dismiss one three times and the theme is retired.

Start free — 2,000 credits How the desk pitches ↓

Suggestions on every plan · credit-metered — the estimate shows before you run

Ideas argued from their knowledge base — not a trending-topics feed.

Fig. 1 — this morning’s pitches · Alma Skin Illustrative slips · the reasons are the product
Taken → calendar
Pitch 01 · LinkedIn essayTheme: barrier science
Why “more moisturiser” is the wrong answer

Why now: core theme, weight 0.8 · no LinkedIn post in 6 days · educate pillar underfed

Pitch 02 · Instagram carouselTheme: routine building
The 3-step evening reset, illustrated

Why now: carousel format unused this fortnight · Instagram slot open Thursday

Pitch 03 · Blog long-formTheme: ingredient literacy
Ceramides, explained like a label-reader

Why now: persona “the ingredient checker” unserved this month

Dismissed ×3 — theme retired
Pitch — · LinkedInTheme: founder diary
A week in the founder’s skincare routine

The desk stops pitching what you keep spiking — that’s real, and it’s the one learning signal we claim.

The desk’s methods — where pitches come from

Argued from the brand brain, filtered like an editor.

Pitches aren’t random and they aren’t trending-topic scrapes — every slip traces back to the client’s own knowledge base and survives four editorial filters before it reaches you.

Method 01 — Themes

The theme map

The knowledge base is distilled into weighted core themes per client — pitches draw from the map, so a skincare brand gets barrier science, not generic “5 productivity hacks”.

Weighted per client · refreshed with the KB
Method 02 — Gap-fill

The calendar tells on itself

The desk scans the client’s calendar for empty weekdays and underfed platforms and pitches straight into the holes — the “why now” on the slip is a real gap, not decoration.

Real weekday-gap logic · fills toward your cadence
Method 03 — Variety rules

No two identical pitches

Editorial filters enforce format diversity and platform rotation — an unused platform gets a boost, near-duplicate ideas get merged, and the pile stays worth reading.

Similarity de-dup · rotation boost · day-of-week bias
Method 04 — Dismissal memory

It learns what you spike

Dismiss a pitch and it’s remembered. Dismiss the same theme three times and the desk retires it — the one learning behaviour here we claim, because it’s the one that exists.

Theme-level suppression at 3+ dismissals
After the presses run — the returns loop

What the engagement data teaches — with the thresholds printed.

Published posts report their engagement back, and three instruments read it. Each one switches on at a printed threshold — because statistics on four posts would be astrology, and we’d rather show you the gate than fake the wisdom.

Instrument 01 — Prediction

Engagement bands, not promises

Before a post ships, its predicted engagement rate is banded from your own history — hook style, length, day, quality score. A rate, honestly — never a promised reach number.

Switches on at ≈10 published posts with engagement
Instrument 02 — Calibration

The score learns your audience

Real correlation math checks which quality dimensions actually predict engagement for this client — and re-weights the score only when the evidence clears the bar.

Applies at 20+ samples · 0.7 confidence — never sooner
Instrument 03 — Fatigue

Knows when a topic is tired

Post the same keyword often enough and the instrument watches the trend — a real decline flags the topic before the audience files the complaint.

Reads at 3+ posts per keyword in 90 days

Straight answer: these instruments calibrate the quality score and your judgement — they don’t currently steer which topics the pitch desk deals. Two desks, side by side. If we wire them together, the changelog says so first.

The scoring gate itself lives on the AI Pipeline page
Said plainly — how we keep this honest
On “it learns”

Today the pitch desk learns from one signal: your dismissals. It does not yet learn from what you use, save, or edit — plenty of tools imply otherwise about themselves; we’d rather under-claim and ship the rest later.

On predictions

Predictions are engagement-rate bands from your own history, and they need about ten published posts before they say anything at all. “Learns over time” is the claim — not a crystal ball.

The morning pile is dealt

Take tomorrow’s pitch.

Start free with 2,000 credits — paste a client’s URL and the desk starts pitching from their world the same day. Roughly a week of real production, on us.

Start with 2,000 free credits See the rate card

No card · No trial countdown · No per-client fees

luminarforge.
LinkedIn-native today · Instagram & Facebook rolling out
Features
AI Pipeline Voice DNA Campaigns Templates AI Imagery Publishing SEO Hub AI Suggestions
This page
The desk The returns loop The rate card Compare
Catalogue
The print room Pricing Trust & security Start free Log in
Company & legal
Contact Terms Privacy Cookies

The pitch desk deals suggestions from each client’s own knowledge base. Pitch slips shown are illustrative; the theme map, gap-fill, variety filters, dismissal memory and the three returns instruments with their printed thresholds reflect the implementation as shipped. The returns instruments calibrate the quality score — they do not currently steer which topics get pitched, and this page says so. © 2026 Luminar Works.